I've been reading about about especially the RadGrid (I am a customer of yours) and have notices you mentioning the awesome performance of the RadGrid which is great. Now I am thinking, how should I as a developer make use of this in the best way, ie, how is it intended to be used. This is of course a question with many answers, but as far as I can see it, the whole idea, unless you have absurd amount of data, or maybe handle very large "blobs", must be that instead of writing stored procedures etc. that does the searching/filtering for you, you should load all the data into the RadGrid, and then let it do it's magic. What I mean is this, if you do a lot of filtering etc. before the data reach the RadGrid, much of the whole advantage with the great performance would be lost, wouldn’t it?
What I am thinking is kind of like this, making stored procedures (or similar) that takes a lot of parameters and does filtering/searching for you isn’t really that difficult, but it takes a lot of time to do, and then you have to make search forms for all these in your web pages on top of that. It would be much quicker if I could just say "Hey, the RadGrid says that it has great performance, and it also offers great filtering capabilities, let's use that instead and save a lot of time!". Is this the intended usage you had in mind when putting in a lot of effort in optimizing the RadGrid performance? Do you have some recommendations, best practices or similar on how to use it, when not to use it etc? Have you done any benchmarking or similar that I as a web developer can use when my DBA comes knocking on the door screaming about that I load all the data from the database?
I hope you understand my questions here, and I know there isn’t an exact answer to this, and that isn’t really what I am looking for, but rather more guidelines on how to use this in a good way!